@ Montanna - their sample size is huge, which certainly helps. But that is only half of it. I am not convinced that their maths is correct. The fact that no-one knows their methodology (ie do they deliberately undertrack a game if they overtracked last month) certainly doesn't help me believe them







